The review is notable for its methodological discipline. Where much of the existing literature takes a broad, sweeping approach to AI in schools, the Greek team narrowed its scope deliberately to the underexplored niche of lesson planning. The analysis follows the search–appraisal–synthesis–analysis framework, a structured procedure for identifying relevant studies, judging their quality, and extracting coherent findings from a heterogeneous body of work. The 53 studies were then organized around four central themes: the quality and pedagogical value of AI-generated plans, the challenges and limitations of the technology, teachers’ attitudes and perceptions, and proposed improvements and future directions. This thematic structure allows the authors to draw a clearer picture than previous, more diffuse surveys of the field.

The headline finding is a paradox that will resonate with any teacher who has experimented with ChatGPT, Google Gemini, or Microsoft Copilot. Large language models are genuinely good at helping educators organize content and save time. They can produce a complete, well-formatted lesson plan in seconds, complete with learning objectives, staged activities, and differentiation suggestions, a task that might otherwise consume an evening. Yet the literature consistently shows that this output often lacks the flexibility, contextual awareness, and pedagogical depth needed for actual classroom use. The plans look right on the page; whether they work with thirty restless adolescents on a Friday afternoon is another matter entirely.

The technical reason for this shortfall lies in how large language models work. These systems generate text by predicting likely continuations based on patterns in their training data, not by drawing on lived teaching experience or knowledge of a specific group of learners. The result, documented across the reviewed studies, is output that tends toward the generic. A model asked to plan a physics lesson for junior high school students cannot know that this particular class has already struggled with a prerequisite concept, or that the school lacks the equipment a hands-on activity requires. Studies cited in the review, including comparative analyses of ChatGPT and Gemini in physics education and content analyses of AI-generated mathematics plans, repeatedly identify this absence of situational judgment as the technology’s central weakness.

Contextual blindness extends to culture and history as well. Research on foreign language lesson plan creation has documented trends, variability, and historical biases in chatbot output, suggesting that the plans a model produces may quietly encode assumptions about curriculum, pedagogy, and even politics drawn from skewed training data. Work in Japanese teacher education has examined the social, cultural, and political dimensions of relying on generative AI in lesson planning, while studies of AI-generated English lesson plans for students with intellectual disabilities raise pointed questions about whether these tools can serve learners with specialized needs. A lesson plan is never a neutral document, and the review makes clear that outsourcing its drafting does not outsource its values.

Perhaps the most striking strand of evidence concerns what happens when humans and machines are compared directly. In one study of music education, teachers were asked to label lesson plans as AI-generated or human-made and to rate their quality, with results revealing how difficult that discrimination has become and how quality judgments vary. Other experimental work has asked whether lesson plans created by ChatGPT are more effective than teacher-designed ones, with mixed outcomes. Meanwhile, studies applying frameworks such as Universal Design for Learning to AI-generated plans have found the output serviceable but shallow, prompting one research team to title its analysis with the pointed question of whether AI-generated plans are better than nothing. The overall picture is of tools that clear a low bar convincingly but have not yet demonstrated superiority over experienced professionals.

Teachers’ own attitudes, the review’s third theme, are more nuanced than either hype or panic would suggest. Across studies of pre-service and in-service teachers in mathematics, science, English language teaching, social studies, and vocational education, educators consistently acknowledge the time savings and organizational help that chatbots provide. Many teacher education programs have begun integrating these tools directly, using ChatGPT as a reflection tool to promote the lesson planning competencies of trainee teachers, or as an assistant with which preservice secondary mathematics teachers rehearse their planning. Yet the same studies emphasize, almost without exception, the importance of teacher involvement in reviewing, adapting, and critically evaluating whatever the model produces. The consensus position is not replacement but supervision.

The review also documents a rapidly evolving effort to make the tools themselves better. Computer scientists have proposed systems such as LessonPlanner, designed to help novice teachers produce pedagogy-driven plans with large language models, and generation pipelines that use self-critique prompting, in which the model evaluates and revises its own draft plans. Retrieval augmented generation has been prototyped for Ugandan secondary schools facing a new national curriculum, grounding model output in authoritative local documents rather than relying on parametric memory alone. Fine-tuned assistants tailored specifically to lesson planning have been tested with teachers to compare perceptions and use. These engineering approaches attack the technology’s weaknesses directly, though the review suggests they remain early-stage.

Against these advances, the authors identify a significant shortfall in the existing literature: there is still limited empirical evidence directly comparing lesson plans designed by teachers with those produced by large language models in controlled settings. Much of the 53-study corpus relies on perceptions, small case studies, or qualitative analysis rather than head-to-head experimental comparisons measuring student outcomes. Until such evidence accumulates, claims that AI can plan lessons as well as or better than teachers rest on shaky ground. The review’s authors argue that as these tools become more common, the field needs practical research, thoughtful integration into teaching practice, and ongoing professional development to ensure responsible and effective use in education.

The broader lesson of the review is about the division of labor between human expertise and machine fluency. Large language models excel at producing well-structured text quickly, and lesson plans are, on their surface, exactly that. But effective teaching depends on knowing learners, adapting in real time, and exercising professional judgment, capacities that no current model possesses. The literature synthesized by the University of Thessaly team points toward a future in which AI handles the scaffolding of planning while teachers supply the pedagogical depth, contextual sensitivity, and critical oversight that turn a plausible document into a good lesson. That future is not automatic; it depends on the empirical research and teacher training the review calls for, and on educators who treat every machine-drafted plan as a first draft to be interrogated rather than a finished product to be delivered.

Subject of Research: The role of large language models in generating and evaluating lesson plans in education

Article Title: Evaluating the Role of Large Language Models in Lesson Planning: Insights from a Narrative Review

Article References: Failadis, V. A., Tasoulis, S. K., Georgakopoulos, S. V., & Plagianakos, V. P. (2026). Evaluating the Role of Large Language Models in Lesson Planning: Insights from a Narrative Review. Frontiers of Digital Education, 3(3), Article 25. https://doi.org/10.1007/s44366-026-0099-6

Image Credits: AI Generated

DOI: 10.1007/s44366-026-0099-6

Keywords: large language models, lesson planning, ChatGPT, generative AI, teacher perceptions, pedagogical quality, narrative review, teacher education, AI in education, instructional design, retrieval augmented generation, pre-service teachers

The study was conducted by Vassilis A. Failadis, Sotiris K. Tasoulis, Spiros V. Georgakopoulos, and Vassilis P. Plagianakos of the University of Thessaly in Greece, and appeared online on September 11, 2026. Rather than surveying the sprawling field of artificial intelligence in education at large, the authors deliberately narrowed their focus to lesson planning, an essential component of a teacher’s daily workload and a core element of professional preparation. Following the search-appraisal-synthesis-analysis framework, known as SALSA, the researchers examined 53 recent publications and organized their findings around four themes: the quality and pedagogical value of AI-generated plans, the challenges and limitations of the technology, teachers’ attitudes and perceptions, and possible improvements and future directions. This focused lens matters, because lesson planning sits at the intersection of content knowledge, curriculum requirements, classroom context, and student needs, making it a revealing test case for what generative models can and cannot do.

The technical verdict on output quality is nuanced. Across the reviewed studies, large language models are frequently praised for producing well-organized plans with clear instructional formats. Platforms such as ChatGPT can reliably follow common instructional models, arranging activities into recognizable stages such as brainstorming, group discussion, guided exploration, presentation, and evaluation. In structural terms, the models have absorbed the conventions of lesson design from their training data and reproduce them fluently. Yet the empirical evidence consistently shows that these plans provide only a general framework and require substantial adaptation before they can function in a real classroom. The plans look like lessons; they do not yet behave like lessons, because they lack the embedded awareness of particular students, curricula, and teaching conditions that experienced educators bring to every plan they write.

Four recurring limitations emerge from the literature with striking consistency. The first is limited pedagogical depth: the teaching practices suggested by large language models tend to promote procedural knowledge rather than conceptual understanding, offering little support for complex learning processes such as reasoning, reflection, and justification. The second is weak differentiation, since the plans rarely include strategies for individualization or alternatives for students with different learning profiles, a shortcoming that becomes especially pronounced in specialized contexts such as special education. The third is unrealistic time planning. One study found that a plan designed by an AI for a single hour would require at least three hours in practice, with the models assigning very short durations to essential tasks while allocating generous time to reminder stages. The fourth is inconsistent output quality, with high variance reported even within identical prompt groups; outputs generated from the same prompt were found to differ by at least five rubric elements, a level of variability that undermines reliability in any professional workflow.

These findings lead the authors to an important practical conclusion about how teachers should interact with the technology. Producing a usable plan from a large language model is not a one-step process but an iterative one, in which the quality of the generated material depends partly on the clarity and specificity of the prompts. This reframes lesson planning with AI as a skill in its own right, blending prompt engineering with pedagogical expertise. A vague request yields a generic skeleton; a precise, context-rich prompt can yield something considerably more useful. But even the best outputs remain drafts that demand professional refinement, because no prompt can fully encode the tacit knowledge a teacher accumulates about how a particular class learns, stalls, and responds.

The review also paints a detailed picture of how teachers themselves perceive these tools. Most educators acknowledge the usefulness of artificial intelligence in lesson planning while simultaneously expressing reservations about its pedagogical adequacy. Teachers commonly use AI to generate content, yet many still prefer to work collaboratively with colleagues and emphasize the need for clear institutional policies to regulate its use. Some see the models as a helpful aid but do not believe they can replace a teacher’s pedagogical judgment; many agree that AI can assist with the initial organization of a lesson while stressing that the final responsibility for adapting and applying the plan lies with the teacher. Intriguingly, the authors observed a generational divide: younger educators tend to be more open to using AI, while more experienced teachers remain cautious, a difference the researchers attribute to the influence of professional experience and familiarity with classroom practice on teachers’ perspectives.

To move the field forward, the review identifies four main directions that appear repeatedly in the literature. The first is teacher training and AI literacy, ensuring that educators understand what these systems can actually do and develop the ability to adapt and assess the content they produce. The second is prompt engineering and reflective use, since the pedagogical value of generated content depends directly on how clear, precise, and well-targeted the prompts are. The third is policy development and ethical guidelines, with clear rules needed to guarantee responsible and educationally appropriate deployment of AI tools in schools. The fourth is model improvement and pedagogical alignment, calling for enhanced features and greater flexibility in instructional design, including guided interfaces and options that allow teachers to supply contextual information about their students, curriculum, and constraints. Together, these directions sketch a roadmap in which the technology and the profession evolve in tandem rather than in competition.

The review is equally candid about what the research itself still does not know, identifying four significant gaps. Most existing studies remain theoretical or rely on teachers’ opinions without examining what happens when AI-generated plans are actually applied in classrooms. Most are built on qualitative observations, interviews, or questionnaires, with limited direct comparisons between teacher-created and machine-generated plans. The literature is also geographically concentrated, with the United States and Türkiye appearing most frequently, raising questions about how findings transfer to other educational systems. Finally, little attention has been paid to the processes through which teachers interact with AI systems during planning, leaving the human-machine workflow largely unexplored. Recent empirical work nonetheless offers telling hints: one study found that although teacher-created plans were rated higher in quality overall, teachers were unable to reliably distinguish them from AI-generated ones, while another found that refined AI-generated mathematics lesson plans received high evaluation scores and in several cases outperformed teacher-created plans, even though teacher-generated procedures aligned more closely with actual classroom practice. The gap between perceived and actual quality is clearly a live scientific puzzle.

The overall conclusion is measured rather than sensational. Large language models can serve as useful tools for lesson planning, but they are not yet capable of replacing the teacher’s pedagogical role. Their outputs are usually well-structured, organized, and goal-oriented, yet they tend to fall short in adaptation, differentiation, and pedagogy, the dimensions that determine whether a plan actually teaches anyone anything. As the authors put it, these systems should currently be viewed as support for lesson planning rather than as ready-made planning tools, and should not yet be considered a substitute for pedagogical expertise; their use makes sense when they function as support tools rather than as replacements for professional judgment. For a technology often portrayed as poised to automate teaching outright, the evidence points to a more grounded future: AI as a fast, tireless drafting assistant, and the teacher as the indispensable editor who knows exactly which students are sitting in the room.

Subject of Research: The use of large language models for lesson planning in education and their pedagogical limitations

Article Title: Large language models in lesson planning: Useful aid, but not yet ready to replace teachers

Article References: Large language models in lesson planning: Useful aid, but not yet ready to replace teachers. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: large language models, lesson planning, generative AI, ChatGPT, education technology, pedagogy, teacher attitudes, prompt engineering, AI literacy, narrative review, differentiation, Frontiers of Digital Education